JARVIS MCP

- Command Execution: Run shell commands on the local system with proper error handling
- File Operations: Read, write, and manage files on the local system
- Directory Visualization: Generate recursive tree views of file systems as JSON structures
- Working Directory Support: Execute commands in specific directories
- Robust Error Handling: Detailed error messages and validation
- Comprehensive Output: Capture and return both stdout and stderr
- Simple Integration: Standard I/O interface for easy integration with various clients

- Go 1.24.1 or higher
- Git (for cloning the repository)

Execute the binary:

``bash

1. Configure the execute_command tool:
- Name: Execute Command
- Description: Execute shell commands on your local machine
- Path: Full path to your jarvis-mcp binary (e.g., /Users/username/jarvis-mcp/out/jarvis-mcp`)
- Arguments: Leave empty (the server uses stdin/stdout)
- Working Directory: Optional; specify a default working directory

2. Save the configuration

You can configure additional tools for specific file operations. For example:

1. Configure the read_file tool:
- Name: Read File
- Description: Read the contents of a file on your system
- Path: Same path as your jarvis-mcp binary
- Arguments: Leave empty

2. Configure the write_file tool:
- Name: Write File
- Description: Write content to a file on your system
- Path: Same path as your jarvis-mcp binary
- Arguments: Leave empty

3. Configure additional tools following the same pattern for:
- list_directory: List directory contents
- create_directory: Create new directories
- move_file: Move or rename files
- search_files: Search for files
- get_file_info: Get file metadata
- directory_tree: Generate a recursive tree view of files and directories

Once configured, you can invoke these tools during conversations with Claude:

1. Type a request like "Please show me the contents of my .bashrc file"
2. Claude will display a tool selection interface
3. Select the appropriate tool (e.g., "Read File")
4. Claude will use JARVIS MCP to execute the operation
5. The results will be displayed in your conversation

1. Configure the execute_command tool:
- Name: Execute Command
- Description: Execute shell commands on your local machine
- Path: Full path to your jarvis-mcp binary (e.g., /Users/username/jarvis-mcp/out/jarvis-mcp)
- Arguments: Leave empty (the server uses stdin/stdout)
- Working Directory: Optional; specify a default working directory

2. Save the configuration

You can configure additional tools for specific file operations. For example:

1. Configure the read_file tool:
- Name: Read File
- Description: Read the contents of a file on your system
- Path: Same path as your jarvis-mcp binary
- Arguments: Leave empty

2. Configure the write_file tool:
- Name: Write File
- Description: Write content to a file on your system
- Path: Same path as your jarvis-mcp binary
- Arguments: Leave empty

3. Configure additional tools following the same pattern for:
- list_directory: List directory contents
- create_directory: Create new directories
- move_file: Move or rename files
- search_files: Search for files
- get_file_info: Get file metadata
- directory_tree: Generate a recursive tree view of files and directories

Once configured, you can invoke these tools during conversations with Claude:

1. Type a request like "Please show me the contents of my .bashrc file"
2. Claude will display a tool selection interface
3. Select the appropriate tool (e.g., "Read File")
4. Claude will use JARVIS MCP to execute the operation
5. The results will be displayed in your conversation

To extend JARVIS MCP with additional functionality, create a new file in the appropriate package following this pattern:

package mypackage
import (

    "context"

    "errors"
"github.com/mark3labs/mcp-go/mcp"

    "github.com/mark3labs/mcp-go/server"

)
func GetMyTool() (tool mcp.Tool, handler server.ToolHandlerFunc) {

    return mcp.NewTool("my_tool",

        mcp.WithDescription("Description of the tool"),

        mcp.WithString("param_name",

            mcp.Required(),

            mcp.Description("Parameter description"),

        ),

    ), myToolHandler

}
func myToolHandler(ctx context.Context, request mcp.CallToolRequest) (*mcp.CallToolResult, error) {

    // Parameter validation

    param, ok := request.Params.Arguments["param_name"].(string)

    if !ok {

        return nil, errors.New("parameter is required")

    }
// Tool implementation

    result, err := doSomething(param)

    if err != nil {

        return nil, err

    }
return mcp.NewToolResultText(result), nil

}

Then register the tool in cmd/jarvis/main.go:

mcpServer.AddTool(mypackage.GetMyTool())